This invention relates to a method and an apparatus for electrical power conditioning and thermal capture/rejection management systems; and more particularly, in one aspect, to integrating electrical power conditioning techniques and heat capture and removal techniques into or onto a common substrate, such as silicon, germanium, gallium arsenic.
Electronic and electrical devices continue to demand additional power as the number of transistors on a semiconductor device, for example a microprocessor, increase dramatically. As a result of that increasing demand, there is an increasing demand on the power conditioning and heat rejection capabilities of systems that support such devices. For example, as microprocessor speed and transistor count increase, there is an increasing requirement for electrical power (an increase in average power consumption) conditioning. Further, as more and more functions are integrated into the microprocessor, for example, the functions typically performed by the floating point processors and video or graphics processors, the power conditioning system must address or respond to the rapidly varying temporal and spatial levels of power consumption. Moreover, the increase in microprocessor speed and transistor count, and the incorporation of more and more functions into the microprocessor, have also created a rapidly increasing requirement to capture and remove heat generated by such microprocessors.
Power supplies are available to meet the power demands, however, the power supply is often located some distance from the consuming device. The finite wire lengths between the supply and the device include capacitance and inductance that introduce time delays in the delivery of power in response to changes in demand by the consuming device. As mentioned above, the temporal change in power consumption of, for example, a microprocessor, is increasing as processor speeds increase and as more and more functions are incorporated into the microprocessor. In response, power conditioning electrical/electronics systems are being placed closer and closer to the consuming device. Locating the power conditioning elements, such as voltage regulators, capacitors, DC-DC converters, near the consuming device may address the concerns regarding the power conditioning needs.
A conventional configuration of the power conditioning system is illustrated in FIG. 1. That system often includes discrete capacitors, voltage regulators, and AC-DC or DC-DC converters. Briefly, discrete capacitors typically are located in physical proximity with and electrically connected to the integrated circuit device. As such, sudden demands by the device during operation may be satisfied by the charge stored on the capacitor, thereby maintaining a relatively constant input voltage for the time necessary for the increased demand to be addressed by the supply. Such capacitors are typically known as bypass capacitors, and are common elements in analog circuit design, digital circuit design, and power device circuit design.
Voltage regulators are employed to take input power at a high voltage (for example, 7 volts), and provide relatively stable output power at a lower voltage (for example, 1 to 5 volts). Voltage regulators tend to provide the lower voltage with greatly increased immunity to variations in the high voltage level, or to variations in current drawn by the consuming device. Regulators are commonly employed in designs of analog and digital electronic power conditioning systems, and are increasingly likely to be placed in proximity to devices that have rapidly time-varying power requirements.
AC-DC and DC-DC converters are employed to transform a particular supply voltage from a convenient source into an appropriate form for consumption by, for example, the integrated circuit device. In many cases, system power electronics provide for a single, relatively high voltage (for example, 48 volt DC, or 110 volt AC), whereas the integrated circuit device may require very different supply voltages (for example, 1 to 5 volts, DC). Under this circumstance, converters transform the power and provide the input voltage required by the device. In some systems, converters are located as close to the consuming device as possible so as to provide stable voltage during variations in power consumption by that device. (See, for example, U.S. Pat. Nos. 5,901,040; 6,191,945; and 6,285,550).
In addition to the power management considerations, the increase in power consumption of these devices has imposed an additional burden on the thermal management system (i.e., systems that capture, remove and/or reject energy in the form of heat). In response, thermal management systems have employed such conventional techniques as heat sinks, fans, cold plates systems that employ cooling water, and/or combinations thereof for heat-capture, removal and rejection from, for example, an integrated circuit device. Such conventional heat management designs locate the thermal capture and rejection elements on or very near the integrated circuit device packaging. (See, for example, U.S. Pat. Nos. 6,191,945 and 6,285,550).
For example, with reference to FIG. 1, heat sinks generally consist of metal plates with fins that transport heat from the consuming device to the surrounding air by natural convection. Heat sinks tend to be located or positioned directly on the integrated circuit device packaging. Heat sinks serve to increase the area of contact between the device and the surrounding air, thereby reducing the temperature rise for a given power.
One technique to enhance the heat transfer between a heat sink and the surrounding air is to employ a fan (typically rotating blades driven by electric motors) in conjunction with a heat sink. Fans may enhance the heat transfer between a heat sink and the surrounding air by causing the air to circulate through the heat sink with greater velocity than by natural convection.
Another technique used by conventional systems to enhance the capabilities of the thermal management system is to reduce the thermal resistance between the consuming device and the heat sink. This often involves reducing the number and thickness of the layers between the device, the device package and the heat sink. (See, for example, U.S. Pat. Nos. 6,191,945 and 6,285,550).
In sum, conventional systems address power conditioning and thermal management requirements by placing both the power conditioning and heat capture and rejection elements as close to the integrated circuit device as possible. This has led to the typical, conventional layout that is illustrated in FIG. 1. With reference to FIG. 1, the consuming device is an integrated circuit device. The thermal management element is heat sink that is in contact with the consuming device. In some implementation, the heat capture, removal and rejection (via the heat sink) may be relatively high.
Further, the power conditioning circuitry (capacitors, voltage regulators, AC-DC and DC-DC converters) is positioned next to the consuming device to reduce the wiring length between the supply and the integrated circuit device.
While such conventional power conditioning and thermal management techniques may be suitable for power consumption and heat capture/rejection requirements for some current device, conventional techniques are unlikely to address the anticipated increases in both power consumption and heat capture, removal and rejection requirements of other current devices as well as future devices. Accordingly, there is a need for new power conditioning techniques to accommodate anticipated increases in both power consumption and heat capture, removal and/or rejection requirements.
Moreover, there is a need for improved power conditioning and thermal management techniques to accommodate increases in both power consumption and heat capture, removal and rejection requirements of current and future devices. Further, there is a need for improved power conditioning and thermal management techniques for devices that may be implemented in space-constrained applications (for example, portable computers). In this regard, there is a need for incorporating the power conditioning and heat capture/rejection elements into the same volume in a stacked configuration as well as address the anticipated increases in both power consumption and heat capture, removal and rejection requirements.
In addition, there is a need for an improved technique(s) of power conditioning and heat capture/rejection that integrate the power conditioning and heat capture/rejection elements with the consuming device (for example, an integrated circuit device) itself—thereby reducing the deficiencies in the power conditioning due to delays in signal propagation and reducing the thermal resistance from the device to the heat sink due to physical separation and additional interfaces. This results in increasing the overall efficiency of both power conditioning and thermal management capabilities of the system.
Moreover, there is a need for power conditioning and heat capture/rejection elements that are stacked in a compact configuration to facilitate a compact packaged device which limits deficiencies in the power conditioning due to delays in signal propagation and enhances the thermal attributes of the packaged device.
Further, while such conventional power conditioning techniques may be suitable for some applications, there is a need for a power conditioning technique that addresses the anticipated increases in power consumption in all applications. For example, there is a need for improved power conditioning techniques for devices that may be implemented in space-constrained applications. Accordingly, there is a need for improved power conditioning techniques to accommodate anticipated increases in power consumption as well as applications having stringent space requirements.